Solving Banking’s Top Data Science Challenges Through Automation

Digital Banking


Ryohei Fujimaki, Ph.D., founder and CEO of dotData

While the technology running and protecting the “back office” of the bank continues to evolve and improve, customer-facing technology often lags behind, reducing loyalty and opening the door for more agile fintech challengers. With a new generation of banking customers demanding multi-channel access and more personalized, 24/7 banking services, banks must find new ways to meet these needs to retain their customers and grow their business.

Banks have always had the benefit of having vast amounts of customer data. This large volume of data can help banks engage with customers more meaningfully — as long as they’re able to collect, interpret, and leverage that data fast enough to drive better products, services and operational efficiencies.

New innovations in AI and machine learning can make this happen, and companies have been investing accordingly. According to IDC, the banking industry will spend $3.3 billion on AI this year, in areas such as cybersecurity and fraud protection, more tailored customer interactions, and faster, more efficient operations.

But while investments in AI are growing, banks are often finding that their existing analytics and business intelligence technology and talent aren’t capable of meeting their current and expanding needs. Challenges in resources, technology infrastructure, and the ability to operationalize models quickly and efficiently can prevent financial institutions from fully leveraging AI and data science to drive business impact.

To overcome these challenges, and maximize the ROI on AI investments, banks must look to innovative solutions such as machine learning and data science automation.

Aging technology infrastructure – Banks and fintech firms were early adopters of data analytics technology, but these older systems can’t handle the volume and complexity of today’s data for advanced predictive analytics. Often for a data science project, data has to be collected from various sources, which is in disparate formats. The traditional approach to data science cannot deal effectively with data collection and preprocessing, and this step can take  months to complete. New platforms that automate this process are scalable and customizable, and can meet current data needs, accepting data from different sources and in different formats, so banks can more quickly analyze the data for model generation.

Highly regulated data – Banks work with highly regulated data and, therefore, need to ensure transparency of their data science process for data security and to certify that all financial regulations are being adhered to. Financial processes and decisions, such as determining who should be approved for a mortgage, conducting a fraud check for a credit card, or determining what the interest rate should be for a given loan, can be exponentially sped up by using AI, but the process needs to be explainable and transparent.  Many data science processes use a black box approach, where the underlying reasoning or processes to produce an output are not available. As such, this type of approach is not ideal for the financial services industry, where it is necessary to understand the influencing variables that are driving a decision. In contrast, white box models, where the influencing variables as well as the mathematical process to process those variables are visible and clearly explained, enable organizations to make data-driven decisions with unprecedented levels of transparency and accountability. Automated, end-to-end data science platforms use white box analytics to provide fully transparent, accurate models that meet regulatory requirements.

Talent and speed to market – Data science is an interdisciplinary domain  that involves many resources from the different functional areas. One project might deploy numerous data engineers, a solution architect, a domain expert, a data scientist (or two), business analysts and perhaps additional resources. Many financial institutions do not have or cannot afford to deploy these resources for a single data science project. Modern, automated data science platforms completely automate the entire data science process, from data collection through production-ready models, including feature engineering. This automation democratizes the process, enabling more participants with different skill levels to effectively execute on projects. This can accelerate the data science process from months to days without the need for additional data science talents.

Financial institutions can leverage automation to overcome the biggest challenges to driving business impact from their data science investments. End-to-end data science automation makes it possible to execute data science processes faster, often in days instead of months, with more transparency. As a result, financial institutions can rapidly scale their AI/ML initiatives to drive transformative business changes.


About Ryohei Fujimaki, Ph.D.

Ryohei Fujimaki is the Founder & CEO of dotData, a spin off of NEC Corporation and the first company focused on delivering end-to-end data science automation for the enterprise. Dr. Fujimaki is a world-renowned data scientist, and was the youngest research fellow appointed in the 119-year history of NEC.